GLACIATION OF THE YAXUTAT BAY REGION 207
belt, vegetation is struggling against increasing odds until, finally, plant growth becomes impossible because of the excessive instability of the soil. The seeds of the alder are taking root all over the ablation moraine within the zone of vegetation; and, near the inner border of that zone, occasional plants may succeed in maintaining growth for a year, or two, or three in especially stable areas; but, ere long, even the most stable soil yields to the undermining and the shrubs perish. One finds plants with green leaves, but with nearly all the roots exposed; others dead; and some, overturned and buried, with their roots in the air and then* branches under ground, sending out roots from the stems. It is a hopeless struggle, and at a distance of a few hundred yards from the inner alder zone few, if any, plants are found. Grasses and annual plants, and even lichens on the larger bowlders no longer find it possible to survive the constant rolling and sliding to which the morainic soil is subjected.
When subjected to advance, as Atrevida, Galiano and Marvine Glaciers were, even the zone of stability temporarily becomes impossible for plant growth. Many trees and bushes are destroyed by the direct thrust and breaking of the ice; but many more are destroyed by the sliding of the soil into the newly-formed crevasses. By these two means great windrows of dead alder and other trees were caused by the 1906 advance.
On the ablation moraine desert, where no plant life can find a foothold, the evidence of instability due to rapid ablation is everywhere present. Here the average depth of moraine cannot be more than two or three feet, and ice is to be seen here and there. The surface is exceedingly hummocky and in whatever direction one travels it is necessary to go up hill and down, rising 50 to 150 feet from valley bottom to hummock crest. There seems no system in detail, though the topography is, in general, a series of roughly circular or elliptical kettles with enclosing ridges of varying heights. From the steeper slopes, bowlders are constantly sliding and when one takes a step on such a slope he may start an avalanche of stones, or he may himself slide down the descent when his foot comes in contact with the thinly-veneered ice slope. Even on the more level parts, bowlders are often perched in such unstable positions on hidden ice pedestals that a mere touch may overturn them. Everywhere instability is evident and there is complete demonstration of the rapid progress of ablation; but when viewed from a distance the ablation moraine seems only a barren waste of debris with little or no ice to be seen.
In spite of the rapid ablation on the moraine-covered ice, there is little flowing water to be seen, excepting where it emerges in great torrents from the glacier margins. There is a great abundance of trickling rills down the steeper slopes, and probably also beneath the moraine on the lesser slopes, but this drainage is usually toward small enclosed basins from which escape is found into the ice. Now and then one of the kettles has no outlet and then a pool or pond is formed, but these are relatively uncommon. The nature of the outlets of the kettles is generally hidden from view by the debris which has slid into them from the margins; but not uncommonly one can hear the falling water as it cascades into hidden moulins. Only in the larger ice valleys are there large moulins. There the ice drainage follows small rills into many small kettles, thence into the glacier and probably soon to its bed. Evidence of the presence of at least some englacial streams was seen in several places where tunnel ends were exposed to view by the melting of the glaciers. Since the sliding of d6bris from the ridges into the kettles transforms areas of thicker moraine to areas of thinner deposit, and since the largest fragments, at least,